My long term goals are (1) to define functional Xist sequences that govern the maintenance of X inactivation by systematic mutational analysis and (2) to deduce the transcriptional control mechanism wherein trans-interactions between the two Xist loci on homologous chromosomes result in only one allele expressing Xist RNA and inactivating the chromosome. I previously generated a line of Xist "knockout" mice and showed that the Xist gene (1) is required in cis for the initiation of X-inactivation, and (2) is essential for the selection mechanism that decides which X chromosome is inactivated. More recently, we have discovered that, in female cells, the Xist allele on the active X chromosome controls the replication timing of the entire inactive wild type X chromosome in trans. Although the inactive X chromosome is normally late replicating, when Xist is deleted from the active X chromosome, the inactive X chromosome becomes early replicating. Since the timing of origin activation has been shown to be controlled by the chromatin environment of each origin of replication, we hypothesize that the chromatin structure of the inactive X chromosome is altered by the active X chromosome. Late replication is a property of heterochromatin and the absence of late replication indicates that the inactive X chromosome has become less heterochromatic. I propose to determine whether a Xist deletion on the active X chromosome destabilizes the maintenance of X-inactivation, (this is suggested by a growth retardation phenotype in Xist+/- females). I also propose to identify the precise structural changes in the early replicating inactive X chromosome. Finally, I propose to precisely localize the functional DNA sequences that control replication timing in trans by introducing an array of mutations into the active X chromosome of female fibroblasts using the cre/Lox-based recombination mediated cassette exchange (RMCE) technique. This technique is highly efficient and can be performed in differentiated cells where the endogenous homologous recombination machinery is inefficient or non-existent.